qualities of good test (rakująca strona)

II. Qua!ities of a good test:

A)    Validity: the test tests what it is svupposed to test (e.g. the test is not vaIid if, when testing writing ability the essay question requires specialist knowledge of history)

1.    Content yalidity: the test should be so constructed as to contain a representative sample of the course, the relationship between the test items and the course objectives always being apparent.

2.    Face yalidity: the test should look as if it is valid (e.g. a test consisting of only three multiple-choice items would not seem valid)

3.    Construct yalidity:    a test has construct validity, it is

capable of measuring certain specific characteristics in accordance with a theory of language behaviour and learning. This type of validity assumes the existence of certain learning theories or constructs underlying the acquisition of abilities and skills.

4.    Empirical (statistical) yalidity:    validity is obtained as a

result of comparing the results of the test with the results of so me criterion measure such as:

an existing test, known or believed to be valid and given at the same time the teacher's ratings or any other such form of independent assessment given at the same time

the subsequent performance of the testees on a certain task measured by some valid test

the teacher's ratings or any other such form of independent assessment given later.

B)    Reliability: _the test should give consistent results. (the time, the

circumstances, scorers, group of testees, version) e.g.    learners of

obviously different levels get roughly the same score on the test; or ifthe same learner gets a different score on two different occasions, the test lacks reliability.

1.    The extent of the sample of materiał selected for testing: the larger the sample, the greater probability that the test as a whole is reliable.

2.    The administration of the test: is the same test administered to different groups under different conditions or at different times?

3.    The number ofsamples of one particular task

4.    Conditions

5.    Persona!factors:

a)    motivation

b)    physical and psychological condition (e.g. illness)

6.    Clarity of instructions: are the various tasks expected from the testees madę elear to all candidates in the rubrics?

7.    The possibility to guess